Qoe-Aware Resource Allocation For Mixed Traffics In Heterogeneous Networks Based On Kuhn-Munkres Algorithm

This paper studies the resource allocation for the downlink transmission in a heterogeneous network with mixed traffics. From the users' side of view, the quality of experience (QoE) is a subjective measurement, based on which the network utility function is defined and taken as a valuation of the system performance. Because the macrocell and picocell are overlaid, the pico base station (PBS) compensates the performance loss of the macrocell by paying for the assigned subcarriers. For the subcarrier allocation, the weighted bipartite graph (WBG) is constructed and a revised Kuhn-Munkres (KM) algorithm is exploited to find the perfect matching. For the power allocation, the first-order derivative of the network utility function is discussed and the optimal power solution is achieved. Simulation results demonstrate that the proposed algorithm always outperforms the average power allocation algorithm and the proportional fairness (PF) algorithm. In addition, this advantage becomes more obvious with the increasing of the cell radius. Moreover, as the cell radius increases, the normalized QoE performance becomes worse.